Food Sci. Biotechnol. 22(6): 1629-1638 (2013)

DOI 10.1007/s10068-013-0260-0

β-Glycosidase-Assisted Bioconversion of Ginsenosides in Purified

Crude Saponin and Extracts from Red Ginseng (Panax ginseng C. A.

Meyer)

Hae-Ryung Park, Kashif Ghafoor, Dohyun Lee, Sunghan Kim, Sun-Hyoung Kim, and Jiyong Park

Received: 14 April 2013 / Revised: 30 May 2013 / Accepted: 30 May 2013 / Published Online: 31 December 2013

© KoSFoST and Springer 2013

Abstract Major ginsenosides in ginseng (Panax ginseng)

and its products are highly glycosylated, hence poorly

absorbed in the gastrointestinal tract. β-Glycosidase-assisted

deglycosylation of pure ginsenosides was peformed to

study bioconversion mechanisms. Ginsenoside standard

compounds, crude saponin, and red ginseng extracts were

incubated with β-glycosidase (0.05% w/v, 55ºC). β-

Glycosidase has a broad specificity for β-glycosidic bonds,

hydrolyzing the β-(1→6), α-(1→6), and α-(1→2) glycosidic

linkages. The final metabolite of protopanaxadiol ginsenosides

was Rg3 while the metabolite of protopanaxatriol ginsenosides

was Rh1. β-Glycosidase treatment of red ginseng extracts

resulted in a decrease in the amounts of Rb1, Rc, Re, and

Rg2 after 24 h, whereas levels of the less glycosylated Rd,

Rb1, Rg, Rg3, Rg1, and Rh1 forms increased. When crude

saponin was incubated with β-glycosidase for 24 h, levels

of Rb1, Rc, Re, and Rg1 decreased while levels of Rd,

Rg3, and Rh1 increased as deglycosylated ginsenosides.

Keywords: ginsenosides, β-glycosidase, bioconversion,

saponin, red ginseng, extract

Introduction

Roots of ginseng (Panax ginseng C.A. Meyer) have been

used as a traditional medicine for preventive and therapeutic

purposes in Asian countries for thousands of years (1).

Ginsenosides are triterpenoid saponins that are found

almost exclusively in ginseng and are believed to be the

main active compounds behind the reported health benefits

of ginseng. Ginsenosides, with few exceptions, share a

similar basic structure containing a dammarane triterpenoidal

skeleton with a modification at C-20 (2,3). Ginsenosides

differ from each other in the types of sugar moieties, their

number, and site of attachment (1). Nearly 50 ginsenosides

have been identified in ginseng (4). The major ginsenosides

in the roots of ginseng are the 20(S)-protopanaxadiol

(PPD) ginsenosides Rb1, Rb2, Rc, and Rd, and the 20(S)-

protopanaxatriol (PPT) ginsenosides Re, Rg1, Rf, and Rg2.

The Re and Rg1 ginsenosides usually account for more

than 90% of the total ginsenoside content of ginseng roots

(5). Ginseng roots are usually processed to red ginseng

using steaming and drying, whereas dried roots (without

the steaming process) are categorized as white ginseng

(1,5). The partly deglycosylated ginsenosides Rh1, Rh2,

and Rg3 are present in red ginseng as artifacts produced

during steaming, along with all the ginsenosides detected

in white ginseng (6). Transformation of ginsenosides into

deglycosylated ginsenosides is required for more effective

in vivo physiological actions (7). For instance, the anticancer

activities of ginsenosides after oral administration are based

on metabolites formed due to deglycosylation by intestinal

bacteria (8). The pharmaceutical activities of the less

glycosylated minor ginsenosides Rd, Rg3, Rh2, and

compound-K are superior to the activities of the major

ginsenosides. These minor ginsenosides are present in

ginseng in small percentages and are produced by

Hae-Ryung Park, Dohyun Lee, Sun-Hyoung Kim, Jiyong Park (�)Department of Biotechnology, Yonsei University, Seoul 120-749, KoreaTel: +82-2-2123-2888; Fax: +82-2-362-7265E-mail: foodpro@yonsei.ac.kr

Kashif GhafoorDepartment of Food Science and Nutrition, King Saud University, Riyadh11451, Saudi Arabia

Sunghan KimNutrex Technology Co., Naju, Jeonnam 520-330, Korea

RESEARCH ARTICLE

1630 Park et al.

hydrolysis of the sugar moieties of the major ginsenosides

(9,10). Once administered orally, ginsenosides are exposed

to gastric juice, digestive, and bacterial enzymes in the

stomach and intestines. However, ginsenosides are not

affected by gastric juice, with the exception of a slight

oxygenation. In addition, the oral bioavailability of intact

ginsenosides from the intestines is extremely low (0.1-

4.4% of Rb1, 3.7% of Rb2, and 1.9-18.4% of Rg1) (8,11)

probably due to a high Mw and hydrophilicity (12).

Different studies have focused on conversion of major

ginsenosides to the more active minor ginsenosides using

heat, mild acid hydrolysis (13), alkaline cleavage (14), and

microbial (15) and enzymatic conversions (10). Chemical

methods result in undesirable side reactions, such as

epimerization, hydration, and hydroxylation (16). In

addition, heat and acid treatments can randomly hydrolyze

glycosidic linkages in other, minor active ginsenosides and

acidic polysaccharides, which can decrease the pharmaceutical

efficacy of ginseng (9). The effects of various microorganisms

on conversion of ginsenosides to their aglycones have been

extensively studied during (4,9,16). However, investigations

of the microbial conversion of ginsenosides have been

limited to only a few ginsenoside types (17-21) and most of

the microorganisms used for transformation of ginsenosides

are not suitable for production of edible products due to

health and safety considerations (16). Moreover, different

studies have used organic chemicals, which are not suitable

for commercial ginseng products, to obtain ginseng

extracts (19), Therefore, enzyme-assisted bioconversion to

hydrolyze sugar moieties at specific positions can be used

as an alternative production process for active minor

ginsenosides. β-Glycosidase is an enzyme capable of

catalyzing the bioconversion of plant glycosides to aglycones

(22) via action upon monoglycosides, diglycosides, and

modified glycosides (23). Despite these useful actions, the

effects of β-glycosidase on hydrolysis of ginsenosides have

not been extensively studied.

The objectives of this study were to investigate the

effects of β-glycosidase on deglycosylation of 10 major

ginsenosides in red ginseng, and to determine the

conversional pathways of ginsenosides after β-glycosidase

treatment. The effects of β-glycosidase on both PPD and

PPT ginsenosides were determined either in pure forms or

in complexes, such as crude saponin and red ginseng

extracts. Red ginseng was extracted using water to test the

applicability of a β-glycosidase treatment to the production

process of commercial red ginseng products.

Materials and Methods

Materials Freshly harvested 4 year old Korean ginseng

roots from Keumsan (Korea) were provided by Nutrex

Technology Co., Ltd. (Naju, Korea). β-Glycosidase derived

from Aspergillus fumigatus was purchased from Amano

Enzyme Inc. (Nagoya, Japan). Membrane syringe filters

(13 mm diameter, 0.45 µm pore size) were purchased from

Millipore (Bedford, MA, USA). Ethanol and ether were

purchased from Samchun Pure Chemical Co., Ltd.

(Pyeongtaek, Korea). Butanol was purchased from J. T.

Baker (Phillisburg, NJ, USA). HPLC-grade methanol,

water, and acetonitrile were purchased from Burdick &

Jackson (Muskegon, MI, USA). Ginsenoside standards

(Rb1, Rb2, Rc, Rd, Rg3, Re, Rg1, Rg2, Rf, and Rh1) were

obtained from ChromaDex Inc. (Irvine, Canada).

β-Glycosidase treatment of ginsenosides Each

ginsenoside standard was treated with β-glycosidase and

analyzed using high performance liquid chromatography

(Dionex; Sunnyvale, CA, USA) after enzyme treatment.

Individual ginsenoside standards were dissolved in analytical

grade methanol and diluted with water (analytical grade) to

produce a 250 ppm solution. The solution was mixed with

β-glycosidase (0.05% w/v) and incubated at 55ºC in a

water bath (SB-651; Eyela, Tokyo, Japan) for 6 h. These

reaction conditions were selected based on a previous

study relating to the use of β-glycosidase for bioconversion

of bioactive compounds (24). A 1 mL aliquot from the

solution was taken every 1 h and analyzed using HPLC. A

control sample was also taken at 0 h just after addition of

β-glycosidase.

Preparation of red ginseng extracts Fresh ginseng

roots were washed with tap water and steamed at 98ºC for

3 h using an autoclave (SS-325; Tomy Kogyo Co., Tokyo,

Japan). Steamed roots were dried at 60ºC for 3 days using

a forced convection type dryer (OF-22GW; Jeio Tech Co.,

Daejeon, Korea) to form red ginseng after Maillard

reactions (25). Dried red ginseng was extracted with water

for 3 h at 80ºC 3 times. Extracts were concentrated using a

rotary evaporator (N-1000; Eyela) under vacuum at 70ºC

until the sugar content of the extract was 70oBx.

β-Glycosidase treatment of crude saponin For comparison

of a ginsenoside bioconversion to a ginsenoside complex,

crude saponin was purified (26) from a red ginseng extract

and treated with β-glycosidase. An amount of 7 g of red

ginseng extract was transferred into a 250 mL round

bottom flask and evaporated using a rotary evaporator (N-

1000; Eyela) under vacuum at 70ºC. The dried residue was

extracted three times with 50 mL of water saturated

butanol at 80ºC for 1 h. The butanol solution was filtered,

transferred to a separatory funnel, and washed with 20 mL

of distilled water to remove impurities. Afterwards, the

butanol solution was transferred to a round bottom flask for

evaporation. The dried residue was extracted with ether at

β-Glycosidase in Bioconversion of Ginsenosides 1631

36ºC for 30 min and cooled for 30 min. After cooling,

ether was discarded and the flask was dried at 105ºC for

20 min, cooled in a desiccator for 30 min, and weighed

until a constant weight was obtained. The weight was then

compared to the weight of the empty flask. The difference

in weights corresponded to the crude saponin extracted

from the red ginseng extract. Purified crude saponin was

dissolved in distilled water at the ratio of 1/200 (w/v), then

mixed with β-glycosidase (0.05% w/v). A β-glycosidase

treated crude saponin solution was incubated at 55ºC in a

water bath (SB-651; Eyela) for 24 h. A 20 mL aliquot of

the solution was taken every 4 h, transferred to a round

bottom flask maintained until a constant weight was

achieved, and evaporated. The dried crude saponin was

dissolved in methanol for HPLC analysis. A control

sample was taken at 0 h after enzyme addition.

β-Glycosidase treatment of red ginseng extracts An

amount of 20 g of a red ginseng extract was dissolved in

180 mL of distilled water and mixed with β-glycosidase

(0.05% w/v) to achieve a homogenous solution. This

solution was placed at 55ºC in a water bath (SB-651;

Eyela) for 24 h. A 20 mL sample was taken at 0 h as a

control, and thereafter every 4 h for further analysis.

Samples were transferred to a 250 mL round-bottom flask

and evaporated using a rotary evaporator (N-1000; Eyela)

under vacuum at 70ºC. The dried residue was extracted

three times with 50 mL of water saturated butanol at 80ºC

for 1 h. The proceeding steps for purification of crude

saponin were the same as described above. Purified crude

saponin was dissolved in methanol before HPLC analysis.

HPLC analysis of ginsenosides The 10 ginsenosides

Rb1, Rb2, Rc, Rd, Rg3, Rg1, Rg2, Re, Rf, and Rh1 were

analyzed using a chromatographic method (27). The HPLC

system consisted of a P680 pump, an ASI-100 auto-

sampler, a UVD340U UV detector, a TCC100 column

oven (Dionex) and a Capcell Pak C18 MG column

(4.6 mm×250 mm; Shiseido, Tokyo, Japan). The detection

wavelength was set at 203 nm and the solvent flow rate

was kept constant at 1.6 mL/min. The mobile phase used

for separation consisted of solvent A (water) and solvent B

(acetonitrile). The gradient elution procedure was 0-10 min

with 80% A, 10-40 min with 68% A, 40-48 min with 58%

A, 48-50 min with 0% A, 50-60 min with 0% A, 60-

62 min with 80% A, and 62-70 min with 80% A. The

injection volume was 20 µL for analysis and all samples

were filtered using 0.45 µm membrane syringe filters

(Millipore) before analysis. Peak identifications were based

on the retention times and comparisons with calibration

curves obtained using injected standard samples prepared

by dissolving pure ginsenosides in analytical grade methanol.

Statistical analysis Experiments and analytical measurements

were carried out in triplicate. Data are represented as

mean±standard deviation (SD). Statistical analysis was

performed using the GraphPad Prism 5.0 software

(GraphPad Software Inc., La Jolla, CA, USA). Data were

analyzed using a one-way analysis of variance (ANOVA).

Significance was defined at p<0.05 and significant results

were further analyzed using Tukey’s post-hoc test.

Results and Discussion

Effects of β-glycosidase on ginsenoside compounds

The protopanaxadiol (PPD) ginsenosides Rb1, Rb2, and

Rc were converted to Rd, which was subsequently

converted to Rg3. Rb1 was almost completely converted to

Rd using a β-glycosidase treatment in 1 h (Fig. 1A). Rd

was further hydrolyzed to Rg3; however, the conversion

rate of Rb1 to Rd was much faster than the conversion rate

of Rd to Rg3 as Rd was detected even after 6 h. Rb2 (Fig.

1B) and Rc (Fig. 1C) were also converted to Rd and Rg3.

However, their conversion patterns were different from the

pattern of Rb1. Both were detected even after 6 h of

enzyme treatment, and the concentration of Rg3 kept

increasing, suggesting that β-glycosidase can hydrolyze not

only the β-(1→6) glycosidic linkage at C-20 of Rb1, but

also the α-(1→6) glycosidic linkage at C-20 of Rb2 and

Rc, producing Rd. β-Glycosidase may release saccharides

in a disaccharide unit, thus producing Rg3 directly from

Rb1, Rb2, and Rc since β-glycosidase is known to act

mainly on disaccharide glycosides (23). The conversion

rate of each ginsenoside was probably dependent on the

affinity of β-glycosidase to the specific ginsenoside. This

enzyme is more efficient in hydrolyzing the β-(1→6)

glycosidic linkage of Rb1 than the α-(1→6) glycosidic

linkages of Rb2 and Rc. Furthermore, hydrolysis of Rc was

faster than hydrolysis of Rb2. Other enzymes may also

produce different rates of ginsenoside bioconversion (10).

The protopanaxatriol (PPT) ginsenosides Re, Rg1, Rf,

and Rg2 were hydrolyzed to Rh1 using a β-glycosidase

treatment. Rg1 was rapidly produced from Re (Fig. 2A)

after 1 h of incubation. After 6 h of incubation, Re was

almost entirely converted to Rg1, which was later partly

converted to Rh1. Similar results were observed for hydrolysis

of Rg1 alone, although most of the Rg1 remained intact

after 6 h of incubation (Fig. 2B). Rf was also partly

hydrolyzed to Rh1 after 6 h of treatment with β-

glycosidase (Fig. 2C). Conversion of Rg2 to Rh1 was

relatively faster than conversion of Rg1 and Rf since Rg2

was completely converted to Rh1 within 6 h (Fig. 2D).

These results indicate that β-glycosidase probably hydrolyzes

the α-(1→2) glycosidic linkage at C-6 of Re and Rg2, and

1632 Park et al.

probably favors the α-rhamnopyranosyl β-(1→2) gluco-

pyranosyl bond of Re and Rg2, compared to α-

glucopyranosyl β-(1→2) glucopyranosyl linkage of Rf.

Zhao et al. (28) observed that the fungal enzymes of

Cylindrocarpon destructans were able to transform the

PPD ginsenosides Rb1, Rb2, Rc, and Rd, whereas the PPT

ginsenosides Re and Rg1 remained unaffected after 48 h of

reaction with these fungal enzymes. The bioconversion

products from PPD ginsenosides in this study included Rd,

ginsenoside XVII, compound O, compound Mb, and

ginsenoside F2, suggesting that enzymes from C.

destructans had a higher specicity for the glycosidic

linkage at C3 than the glycosidic linkage at C20 (28). On

the other hand, Wang et al. (24) reported that a ginsenosidase

type IV from an Aspergillus strain hydrolyzed the PPT

ginsenosides Rg2 and Rf to Rh1, and further hydrolyzed

Rh1 to aglycone, while the enzyme was not able to

hydrolyze PPD ginsenosides. In the present study, β-

glycosidase showed broader specificities than the fungal

enzymes as it transformed both PPD and PPT ginsenosides.

Based on these findings, conversional pathways for PPD

and PPT ginsenosides using a β-glycosidase treatment can

be proposed. The final product of β-glycosidase-assisted

bioconversion of PPD ginsenosides was Rg3, whereas the

final product of PPT ginsenosides was Rh1 (Fig. 3). Rb1,

Rb2, and Rc may either be converted to Rg3 directly, or

first to Rd, which is subsequently converted to Rg3 in the

presence of β-glycosidase at a low concentration with

moderate heating. The 2 β-D-glucopyranosyl moieties are

hydrolyzed from Rb1 to form Rg3, and α-L-arabino-

pyranosyl and β-D-glucopyranosyl are cleaved from Rb2 to

form Rg3. Rc is converted to Rd due to removal of 1 α-D-

arabinofuranosyl, and subsequent removal of 1 β-D-

glucopyranosyl from Rd via β-glycosidase hydrolysis

results in Rg3 (Fig. 3A).

Among PPT ginsenosides, Re is first converted to Rg1

via hydrolysis of the α-L-rhamnopyranosyl moiety, then a

β-D-glucopyranosyl moiety is further removed to form

Rh1. Conversion of Rf and Rg2 to Rh1 requires hydrolysis

of β-D-glucopyranose and α-L-rhamnopyranose, respectively.

Hence no intermediate is produced in these bioconversions

(Fig. 3B).

Rg3 and Rh1 probably have more bioavailability,

compared to the parent ginsenosides, due to lower Mw

(24). In vivo studies have shown that Rg3 can be

metabolized to Rh2 and/or 20(S)-PPD by human intestinal

bacteria (29). This is consistent with a pharmacokinetic

study in rats showing that intra-gastric administration of

10 mg/kg of Rg3 resulted in plasma concentrations of

approximately 40 and 100 ng/mL of Rh2 and 20(S)-PPD,

respectively, 4 h after administration (30). Rh2 and 20(S)-

PPD are probably more physiologically active than the

intact ginsenosides Rb1, Rd, Rb2, and Rc. Rh2 and 20(S)-

PPD have been shown to be more cytotoxic against tumor

cells, compared with the intact ginsenosides (29). However,

intestinal bacterial strains may vary from individual to

individual, and the capacity to metabolize ginsenosides to

more bioactive forms may also vary (9). Therefore,

transformation of ginsenosides prior to ingestion could be

beneficial to maximize absorption and pharmacological

activities. For example, transformation of PPT ginsenosides

into Rh1 using β-glycosidase is expected to increase the

bioactivity of the intact ginsenosides and to aid in further

transformation to ginsenoside F1, and then finally to 20(S)-

Fig. 1. Chromatograms showing conversion of theprotopanaxadiol ginsenosides Rb1 (A), Rb2 (B), and Rc (C)using β-glycosidase treatment. Ginsenoside solutions (250 ppm)were incubated with β-glycosidase (0.05% w/v) at 55ºC in a waterbath for 6 h.

β-Glycosidase in Bioconversion of Ginsenosides 1633

PPT by intestinal bacteria (7). It is also recommended to

treat ginsenosides with β-glycosidase prior to in vivo trials

to assess the health effects. Hence, it should be feasible to

develop a specific bioconversion processes to obtain

specifically designed functional products using appropriate

combinations of ginsenoside substrates and suitable

enzymes (21).

Effects of β-glycosidase on red ginseng extracts Red

ginseng, which is commonly sold in the form of a concentrated

extract, is preferred over white ginseng because of more

health benefits due to increased bioavailability of phenolic

compounds and ginsenosides (31). Although different

studies have reported microbial enzyme treatments of ginseng

extracts for producing compound K and ginsenoside

aglycon (19), all such studies used organic solvents in

preparation of ginseng extracts. Therefore, to evaluate the

industrial applications of β-glycosidase treatments, red

ginseng extracts prepared using water were treated with β-

glycosidase for 24 h and changes in 10 ginsenosides were

analyzed every 4 h. Levels of each ginsenoside were

normalized to relative compositions by comparison to the

total ginsenoside content (sum of 10 ginsenosides) at each

time interval to compensate for sampling error (Table 1).

The levels of the PPD ginsenosides Rb1, Rb2, and Rc

decreased over time due to β-glycosidase treatment, but

this decrease was not significant. The relative proportion of

Rc was significantly (p<0.05) decreased by β-glycosidase

treatment from a beginning concentration of 17.37% to

12.94% after 12 h, 11.43% after 16 h, 10.72% after 20 h,

and 8.64% after 24 h. The content and relative proportion

of Rd was significantly (p<0.05) increased by treatment

with β-glycosidase for 24 h from 0.77 mg/g (8.92%) to

2.37 mg/g (25.99%). A significant increase in the relative

proportion of Rd was evident even after 12 h of treatment.

The level of Rg3 was increased from 0.81 to 1.65% after

24 h due to hydrolysis; however the increase was not

consistent.

Fig. 2. Chromatograms showing the conversion of the protopanaxatriol ginsenosides Re (A), Rg1 (B), Rf (C), and Rg2 (D) usingβ-glycosidase treatment. Ginsenoside solutions (250 ppm) were incubated with β-glycosidase (0.05% w/v) at 55ºC in a water bath for 6h.

1634 Park et al.

Fig. 3. Suggested conversional pathways of protopanaxadiol ginsenosides (A) and protopanaxatriol ginsenosides (B) using β-glycosidase treatment. Glc, β-D-glucopyranosyl; Rha, α-L-rhamnopyranosyl; Ara(f), α-D-arabinofuranosyl; Ara(p), α-L-arabinopyranosyl

β-Glycosidase in Bioconversion of Ginsenosides 1635

Among PPT ginsenosides, the levels of Re and Rg2

showed significant decrease from 1.59 to 0.15 mg/g, and

from 0.19 to 0.01 mg/g, respectively, after 24 h of incubation

with β-glycosidase. The relative proportion of Rg1 increased

from 7.72 to 20.42% and to 20.92% after 20 and 24 h,

respectively, due to conversion of Re to Rg1. The relative

proportion of Rh1 increased from 1.45 to 4.13% after 16 h

of incubation with β-glycosidase. The level and relative

proportion of Rf did not show significant change.

The sum of Rb1 and Rg1 (Rb1+Rg1) is a critical

parameter for commercial red ginseng products in Korea.

According to the Korean Food Code (32), red ginseng

powder or extract must contain 0.8-34 mg/g of Rb1+Rg1.

For immune-stimulating and invigorating purposes, 3-

80 mg of Rb1+Rg1 should be ingested, while 2.4-23 mg of

Rb1+Rg1 should be ingested to inhibit platelet aggregation

(31). The results of this study have shown that the

Rb1+Rg1 amount increased by 130% after 4 h and by

137% after 8 h of incubation with β-glycosidase (Table 1).

The relative proportion of Rb1+Rg1 showed significant

increase in comparison to other ginsenosides after β-

glycosidase treatments for 4, 8, 12, 16, 20, and 24 h. Fast

conversion of Re to Rg1 contributed to the increased level

of Rb1+Rg1, although the level of Rb1 was decreased

slightly over time due to hydrolysis by β-glycosidase. For

incubation times longer than 24 h, the Rb1+Rg1 total

might be decreased due to hydrolysis of Rb1.

Effects of β-glycosidase on crude saponin Crude

saponin purified from red ginseng extracts was incubated

with β-glycosidase (0.05% w/v) at 55ºC for 24 h. Samples

were taken every 4 h over a 24 h period. The levels and

relative proportions of Rb1, Rc, Re, and Rg1 decreased

significantly (p<0.05) after 24 h of treatment with β-

glycosidase (Table 2). Significant decrease in the amount

of Rb1, Rc, and Re were prominent even after 4 h of

enzyme reaction. The level and relative proportion of Rd

started to increase significantly after 4 h of reaction time,

and kept increasing afterwards. The relative proportions of

Rg3, Rg2, and Rh1 also started to show significant increase

after 4 h. Neither the amounts nor the relative proportions

of Rb2 and Rf showed significant changes during 24 h of

enzyme reaction. These results are consistent with results

from the β-glycosidase treatment of individual ginsenosides

since increases in the amounts Rg3 and Rh1 were observed.

Rg3 and Rh1 are the end products in β-glycosidase-assisted

Table 1. Changes in ginsenoside content (mg/g red ginseng extract) and percentage (%) changes in the relative proportions ofindividual ginsenosides in a red ginseng extract using β-glycosidase treatment

Ginsenoside

β-Glycosidase reaction time

0 h 4 h 8 h 12 h 16 h 20 h 24 h

mg/g % mg/g % mg/g % mg/g % mg/g % mg/g % mg/g %

Rb12.8

±0.2732.37±4.42

2.32±0.7

32.69±2.9

2.44±0.44

33.66±3.88

2.36±0.35

32.71±3.8

2.45±0.4

31.26±3.38

2.63±0.28

27.67±0.16

2.37±0.59

26.46±0.11

Rb20.73±0.3

8.57±4.31

0.95±0.39

9.98±5.31

0.89±0.24

9.19±4.52

0.64±0.27

8.76±3.44

0.68±0.21

8.57±2.2

0.77±0.11

8.1±0.39

0.66±0.13

7.4±0.34

Rc1.52

±0.2817.37±1.09

1.54±0.63

15.16±1.7

1.58±0.44

15.08±1.09

0.93±0.08

12.94±0.38*1)

0.9±0.22

11.43±1.3*

1.03±0.21

10.72±1.27*

0.76±0.12

8.64±0.69*

Rd0.77

±0.048.92

±0.711.44

±0.2515.05±3.74

1.68±0.23

16.51±2.39

1.42±0.11

19.82±2.74*

1.73±0.09

22.25±2.75*

2.17±0.22*

22.94±2.44*

2.37±0.91*

25.99±3.27*

Rg30.07

±0.060.81

±0.570.07

±0.040.65

±0.430.13±0.1

1.09±0.79

0.09±0.08

1.19±1.04

0.07±0.07

0.91±0.68

0.12±0.04

1.21±0.24

0.14±0.05

1.65±0.84

Re1.59

±0.3818.17±3.45

0.84±0.57

7.89±3.07*

0.67±0.36

6.15±2.36*

0.25±0.15*

3.37±1.87*

0.17±0.08*

2.15±0.68*

0.16±0.07*

1.64±0.77*

0.15±0.05*

1.64±0.11*

Rg10.69

±0.387.72

±3.611.26±0.7

12.26±5.28

1.32±0.71

12.1±4.58

1.07±0.39

14.96±5.35

1.31±0.55

16.47±5.18

1.94±0.23

20.42±1.09*

1.85±0.31

20.92±1.46*

Rf0.22

±0.012.51

±0.180.28

±0.052.87

±0.530.29±0.1

2.74±0.63

0.18±0.01

2.49±0.18

0.21±0.01

2.76±0.42

0.25±0.05

2.66±0.59

0.25±0.12

2.65±0.66

Rg20.19

±0.092.13

±0.840.09

±0.060.82

±0.28*0.05

±0.030.43

±0.2*0.02

±0.02*0.28

±0.27*0.01

±0.01*0.09

±0.08*0.02

±0.01*0.18

±0.1*0.01

±0.00*0.05

±0.05*

Rh10.13

±0.061.45

±0.520.27

±0.172.58

±1.050.33

±0.173.05

±1.140.25

±0.063.49

±0.620.33±0.1

4.13±0.84*

0.42±0.04

4.45±0.04*

0.4±0.02

4.6±0.79*

Rg1+Rb13.49

±0.6540.09±1.31

4.46±1.4

44.96±3.05*

4.76±1.15

45.76±0.71*

3.43±0.74

47.67±1.65*

3.76±0.95

47.73±1.81*

4.57±0.51

48.09±1.25*

4.22±0.91

47.38±1.36*

1)*Significantly different from the control at 0 h (p<0.05) within a row. Values are expressed as the mean±SD (n=3).

1636 Park et al.

deglycosylation of pure PPD and PPT ginsenosides,

respectively. Furthermore, crude saponin may be a more

favorable substrate for enzyme reaction, compared with

original red ginseng extracts, due to fewer impurities after

the purification processes because ginseng contains

saponins, polysaccharides, and flavonoids (33).

The initial conversion rate of Rb1 was fast so that the

level of Rb1 decreased from 54.54 to 6.68 mg/g after 4 h,

and then decreased to 2.26 mg/g after 24 h. A marked

decrease (34.21 to 5.03 mg/g) in the Rc amount was

observed after 16 h. The level of Re was decreased by 65%

during the first 8 h of incubation, and then decreased

gradually thereafter. The decrease in reaction rates might

be due to saturation of β-glycosidase with an excess of

substrate, or there might not be sufficient available

substrate for enzymatic hydrolysis due to the fast initial

reaction. The level of Rd was increased rapidly after 4 h of

incubation; however, it did not show significant change

afterwards, probably due to β-glycosidase saturation with

substrate, or some of the newly synthesized Rd was further

hydrolyzed to Rg3.

The level of Rg1 started to show significant decrease

(from 10.02 to 5.80 mg/g) after 20 h of reaction. The

relative contents of Rg3 and Rh1 to the total ginsenoside

content showed significant increases after 4 h of incubation

with β-glycosidase. Ko et al. (10) reported synthesis of Rd

from PPD ginsenosides in a crude saponin mixture when

treated with the glycoside hydrolases lactase and β-

galactosidase from Aspergillus oryzae, and cellulase from

Trichoderma viride. Synthesis of Rd and Rg3 from PPT

ginsenosides was also reported in a saponin mixture treated

with lactase from Penicillium sp. β-glycosidase assisted

bioconversion was comparable with other methods of

ginsenoside conversion, such as puffing (use of heating and

pressure). An et al. (26) observed that the Rg3 content of

crude saponin prepared from puffed red ginseng increased

up to 3.35 mg/g after an increase in the puffing pressure.

The β-glycosidase-assisted reaction produced 3.13 mg/g of

Rg3 after 24 h of incubation. Enzymes play important roles

in liberation of ginsenosides from ginseng leaves when

used at an appropriate stage in the extraction process (34).

β-glycosidase favors hydrolysis of the PPD ginsenosides

Rb1, Rc, and Re over the PPT ginsenosides Rg1, Rg2, Rf,

and Rb2. The PPD ginsenosides changed more dramatically

after β-glycosidase treatment in red ginseng extract (Table

1) and crude saponin (Table 2) as discussed earlier. The

Table 2. Changes in ginsenoside contents (mg/g crude saponin) and percentage (%) changes in relative proportions of individualginsenosides in a crude saponin solution using β-glycosidase treatment

Ginsenoside

β-Glycosidase reaction time

0 h 4 h 8 h 12 h 16 h 20 h 24 h

mg/g % mg/g % mg/g % mg/g % mg/g % mg/g % mg/g %

Rb154.54±4.69

31.94 ±2.09

6.68±1.47*1)

5.05±0.35*

3.64±0.92*

3.32±0.41*

3.92±0.56*

2.93±0.35*

4.21±1.33*

3.16±0.27*

3.04±1.27*

3.08±0.93*

2.26±0.86*

2.27±0.48*

Rb224.8

±7.2414.26±2.3

20.41±2.74

15.58±0.59

16.46±3.42

14.98±0.42

20.26±3.73

14.96±0.71

20.68±6.54

15.37±0.5

14.79±6.09

14.27±0.55

15.61±4.64

15.46±1.08

Rc24.21±7.44

13.88±2.11

12.83±2.4*

9.73±0.52*

7.41±1.92*

6.68±0.42*

6.52±1.43*

4.8±0.5*

5.03±1.47*

3.89±0.72*

3.1±1.11*

3.12±0.57*

3.03±1.03*

3.05±0.59*

Rd13.69±2.55

7.94±0.51

53.22±7.34*

40.58±1.00*

49.83±9.53*

45.47±0.88*

64.84±8.99*

48.23±1.88*

65.99±20.83*

49.02±0.77*

52.84±21.42*

50.76±1.79*

51.53±15.38*

51.01±0.95*

Rg31.97

±0.761.13

±0.372.28

±0.371.73

±0.11*2.43

±0.632.19

±0.22*3.1

±0.542.3

±0.11*3.39

±0.982.57

±0.29*3.14

±1.233.07

±0.23*3.11

±1.013.1

±0.33*

Re33.04±5.15

19.46±4.01

17.44±3.81*

13.19±0.96*

12.79±3.33*

11.54±0.65*

14.1±3.04*

10.38±0.83*

11.74±3.45*

8.88±0.61*

8.33±3.13*

8.13±0.6*

7.21±1.63*

7.34±0.71*

Rg110.02±1.43

5.84±0.15

8.68±1.47

6.6±0.3

7.26±1.68

6.59±0.34

8.67±1.73

6.4±0.39

8.27±2.38

6.24±0.3

5.8±2.21*

5.67±0.37

5.46±1.65*

5.46*±0.47

Rf3.01

±0.531.75

±0.082.22

±0.451.69

±0.141.94

±0.521.75±0.11

2.38±0.54

1.75±0.12

2.27±0.68

1.7±0.05

1.8±0.73

1.75±0.07

1.78±0.55

1.77±0.14

Rg23.66

±0.282.15

±0.183.86

±0.582.95

±0.16*3.67

±0.683.35

±0.17*4.79±0.7

3.56±0.26*

5.29±1.68

3.94±0.16*

4.24±1.72

4.09±0.28*

4.28±1.32

4.24±0.2*

Rh12.9

±0.861.67

±0.243.76

±0.352.9

±0.29*4.52

±1.034.12

±0.34*6.35

±1.334.69

±0.33*7.28

±2.71*5.24

±0.75*6.38

±2.916.06

±0.57*6.38

±2.086.3

±0.38*

Rg1+Rb164.56±6.12

37.78±2.25

15.36±2.94*

11.65±0.65*

10.9±2.6*

9.9±0.75*

12.59±2.29*

9.33±0.73*

12.48±3.71*

9.39±0.56*

8.84±3.48*

8.75±1.3*

7.72±2.51*

7.73±0.95*

1)*Significantly different from a control at 0 h (p<0.05) within a row. Values are expressed as the mean±SD (n=3).

β-Glycosidase in Bioconversion of Ginsenosides 1637

PPD ginsenosides Rb1, Rb2, and Rc comprise major

portions of the total ginsenoside content in crude saponin.

Lower amounts of PPT ginsenosides may limit the probability

of interaction with β-glycosidase and, thus, hydrolyzation

by the enzyme. Shehzad et al. (35) reported that the 6

major ginsenosides Rb1, Rb2, Rc, Rd, Re, and Rg1

comprise approximately 90% w/w of the total weight of

saponin, which is consistent with the findings of this report.

At the start of enzyme treatment, these ginsenosides were

present in higher proportions, compared to other ginsenosides

in red ginseng extracts and a purified crude saponin

solution. Conversional pathways for ginsenosides using

various microbial enzymes have been reported (17,19,21).

However, complete transformational pathways were not

developed in these studies, suggesting the presence of

partial pathways for PPD and PPT ginsenosides, and other

ginsenosides. This study is useful for providing conversional

pathways for both PPD and PPT ginsenosides using β-

glycosidase. Further research will, however, be necessary

to arrive at more in-depth conclusions regarding ginsenoside

bioconversion using β-glycosidase.

The effects of β-glycosidase on both PPD and PPT

ginsenosides are shown and activities of the β-glycosidase

enzyme in different settings using pure ginsenosides, crude

saponin, and red ginseng extracts are compared. Determination

of conversional pathways of ginsenosides was accomplished

by incubating ginsenoside standards with β-glycosidase,

followed by HPLC analysis. β-Glycosidase has a broad

specificity for various β-glycosidic bonds, including β-

(1→6), α-(1→6), and α-(1→2)-glycosidic linkages. The

final metabolite of the PPD ginsenosides Rb1, Rb1, Rc,

and Rd was Rg3, while the metabolite of the PPT

ginsenosides Rg1, Rg2, Re, and Rf was Rh1 using β-

glycosidase treatment. β-Glycosidase treatment can raise

the bioactive potential of pure ginsenosides by conversion

to deglycosylated and more active forms. When crude

saponin was incubated with β-glycosidase for 24 h, the

levels of Rb1, Rc, Re, and Rg1 decreased due to hydrolysis

of ginsenoside sugar moieties, while the levels of the

deglycosylated ginsenosides Rd, Rg3, and Rh1 increased.

When β-glycosidase was applied to red ginseng extract, the

levels of Rb1, Rc, Re, and Rg2 decreased after 24 h of

incubation, while the levels of the less glycosylated Rd,

Rg3, Rg1, and Rh1 forms increased. The total amount of

Rb1 and Rg1 (Rb1+Rg1), which is an important criterion

for commercially available red ginseng extracts in Korea,

was increased by 130-140% after β-glycosidase treatment.

Therefore, the β-glycosidase treatment of red ginseng

extract can be an inexpensive and effective way to improve

the quality of red ginseng products with a higher Rb1+Rg1

content. Furthermore, higher amounts of ginsenoside

aglycones in enzyme-treated red ginseng extracts with

increased bioavailability and pharmacological activities of

ginsenosides can benefit consumers.

Acknowledgments The Nutrex Technology Co., Ltd.,

Naju, Korea provided financial support for this research.

References

1. Park JD, Rhee DK, Lee YH. Biological activities and chemistry ofsaponins from Panax ginseng C.A. Meyer. Phytochem. Rev. 4: 159-175 (2005)

2. Huang KC. The Pharmacology of Chinese Herbs. CRC Press, BocaRaton, FL, USA (1999)

3. Shibata S, Tanaka O, Shoji J, Saito H. Chemistry and pharmacologyof panax. pp. 217-284. In: Economic and Medicinal Plants Research.Wagner H, Farnsworth NR, Hikino H (eds). Academic Press,London, UK (1985)

4. Cheng LQ, Na JR, Bang MH, Kim MK, Yang DC. Conversion ofmajor ginsenoside Rb1 to 20(S)-ginsenoside Rg3 by Microbacteriumsp. GS514. Phytochemistry 69: 218-224 (2008)

5. Christensen LP. Ginsenosides chemistry, biosynthesis, analysis, andpotential health effects. Adv. Food Nutr. Res. 55: 1-99 (2009)

6. Kitagawa I, Taniyama T, Shibuya H, Noda T, Yoshikawa M.Chemical studies on crude drug processing. V. On the constituentsof ginseng radix rubra (2): Comparison of the constituents of whiteginseng and red ginseng prepared from the same Panax ginseng.Yakugaku. Zasshi. 107: 495-505 (1987)

7. Tawab MA, Bahr U, Karas M, Wurglics M, Schubert-Zsilavecz M.Degradation of ginsenosides in humans after oral administration.Drug Metab. Dispos. 31: 1065-1071 (2003)

8. Hasegawa H. Proof of the mysterious efficacy of ginseng: Basic andclinical trials: metabolic activation of ginsenoside: Deglycosylationby intestinal bacteria and esterification with fatty acid. J. Pharmacol.Sci. 95: 153-157 (2004)

9. Kim MK, Lee JW, Lee KY, Yang DC. Microbial conversion ofmajor ginsenoside Rb(1) to pharmaceutically active minor ginsenosiderd. J. Microbiol. 43: 456-462 (2005)

10. Ko SR, Suzuki Y, Suzuki K, Choi KJ, Cho BG. Marked productionof ginsenosides Rd, F2, Rg3, and compound K by enzymaticmethod. Chem. Pharm. Bull. 55: 1522-1527 (2007)

11. Hu Y, Luan H, Hao D, Xiao H, Yang S, Yang L. Purification andcharacterization of a novel ginsenoside-hydrolyzing β-D-glucosidasefrom the China white jade snail (Achatina fulica). Enzyme Microb.Tech. 40: 1358-1366 (2007)

12. Karikura M, Miyase T, Tanizawa H, Taniyama T, Takino Y. Studieson absorption, distribution, excretion and metabolism of ginsengsaponins. VI. The decomposition products of ginsenoside Rb2 in thestomach of rats. Chem. Pharm. Bull. 39: 400-404 (1991)

13. Han BH, Park MH, Han YN, Woo LK, Sankawa U, Yahara S,Tanaka O. Degradation of ginseng saponins under mild acidicconditions. Planta Med. 44: 146-149 (1982)

14. Chen YJ, Nose M, Ogihara Y. Alkaline cleavage of ginsenosides.Chem. Pharm. Bull. 35: 1653-1655 (1987)

15. Bae EA, Han MJ, Choo MK, Park SY, Kim DH. Metabolism of20(S)- and 20(R)-ginsenoside Rg3 by human intestinal bacteria andits relation to in vitro biological activities. Biol. Pharm. Bull. 25: 58-63 (2002)

16. Chi H, Ji GE. Transformation of ginsenosides Rb1 and Re fromPanax ginseng by food microorganisms. Biotechnol. Lett. 27: 765-771 (2005)

17. Ko SR, Choi KJ, Suzuki K, Suzuki Y. Enzymatic preparation ofginsenosides Rg2, Rh1, and F1. Chem. Pharm. Bull. 51: 404-408(2003)

18. Wang DM, Yu HS, Song JG, Xu YF, Liu CY, Jin FX. A novelginsenosidase from an Aspergillus strain hydrolyzing 6-O-multi-glycosides of protopanaxatriol-type ginsenosides, named ginsenosidasetype IV. J. Microbiol. Biotechnol. 21: 1057-1063 (2011)

19. Yoo MH, Yeom SJ, Park CS. Production of aglycon protopanaxadiol

1638 Park et al.

via compound K by a thermostable beta-glycosidase fromPyrococcus furiosus. Appl. Microbiol. Biotechnol. 89: 1019-1028(2011)

20. Yu H, Zhang C, Lu M, Sun F, Fu Y, Jin F. Purification andcharacterization of new special ginsenosidase hydrolyzing multi-glycisides of protopanaxadiol ginsenosides, ginsenosidase type I.Chem. Pharm. Bull. 55: 231-235 (2007)

21. Chi H, Kim DH, Ji GE. Transformation of ginsenosides Rb2 and Rcfrom Panax ginseng by food microorganisms. Biol. Pharm. Bull. 28:2102-2105 (2005)

22. Pyo YH, Lee TC, Lee YC. Enrichment of bioactive isoflavones insoymilk fermented with beta-glucosidase-producing lactic acidbacteria. Food Res. Int. 38: 551-559 (2005)

23. Yamamoto S, Okada M, Usui T, Sakada K, Toumoto A, TsuruhamiK. Diglycosidase isolated from microorganisms. U.S. Patent7,109,014 B1 (2006)

24. Ryu YG, Won B, Park HR, Ghafoor K, Park J. Effects of the beta-glycosidase reaction on bio-conversion of isoflavones and qualityduring tofu processing. J. Sci. Food Agr. 90: 843-849 (2010)

25. Ghafoor K, Kim SO, Lee DU, Seong K, Park J. Effects of highhydrostatic pressure on microstructure and color of red ginseng(Panax ginseng). J. Sci. Food Agr. 92: 2975–2982 (2012)

26. An YE, Ahn SC, Yang DC, Park SJ, Kim BY, Baik MY. Chemicalconversion of ginsenosides in puffed red ginseng. LWT-Food Sci.Tech. 44: 370-374 (2011)

27. Kong BM, Park MJ, Min JW, Kim HB, Kim SH, Kim SY. Physico-chemical characteristics of white, fermented and red ginsengextracts. J. Ginseng Res. 32: 238-243 (2008)

28. Zhao X, Gao J, Song C, Fang Q, Wang N, Zhao T, Liu D, Zhou Y.Fungal sensitivity to and enzymatic deglycosylation of ginsenosides.Phytochemistry 78: 65-71 (2012)

29. Bae E, Han MJ, Kim EJ, Kim DH. Transformation of ginsengsaponins to ginsenoside Rh2 by acids and human intestinal bacteriaand biological activities of their transformations. Arch. Pharm. Res.27: 61-67 (2004)

30. Xie HT, Wang GJ, Sun JG, Tucker I, Zhao XC, Xie YY, Li H, JiangX, Wang R, Xu MJ, Wang W. High-performance liquidchromatographic-mass spectrometric determination of ginsenosideRg3 and its metabolites in rat plasma using solidphase extraction forpharmacokinetik studies. J. Chromatogr. B 818: 167-173 (2005)

31. Kim KT, Yoo KM, Lee JW, Eom SH, Hwang IK, Lee CY.Protective effect of steamed American ginseng (Panax quinquefoliusL.) on V79-4 cells induced by oxidative stress. J. Ethnopharmacol.111: 443-450 (2007)

32. KFDA. Food Code. Health Functional Food Division, Food andDrug Administration, Korea. II.2.1.2-1 (2008)

33. Kim DH. Chemical diversity of Panax ginseng, Panax quinquifolium,and Panax notnoginseng. J. Ginseng Res. 36: 1-15 (2012)

34. Lee HJ, Lee HS, Cho HJ, Kim SY, Suh HJ.Utilization of hydrolyticenzymes for the extraction of ginsenosides from Korean ginsengleaves. Process Biochem. 47: 538-543 (2012)

35. Shehzad O, Ha IJ, Park Y, Ha YW, Kim YS. Development of arapid and convenient method to separate eight ginsenosides fromPanax ginseng by high-speed countercurrent chromatographycoupled with evaporative light scattering detection. J. Sep. Sci. 34:1116-1122 (2011)